search

International-Soil-Radiocarbon-Database / ISRaD

Repository for the development and release of ISRaD data and tools
https://international-soil-radiocarbon-database.github.io/ISRaD/
24 stars 15 forks source link

QAQC check for lyr_bot>lyr_top #124

Closed olgavinduskova closed 5 years ago

olgavinduskova commented 5 years ago

It may be useful to have QAQC check that lyr_bot is higher than lyr_top for each layer. (I just found an ingested template with switched lyr_bot and lyr_top for few litter layers, I sent a corrected version to info.israd@gmail.com)

jb388 commented 5 years ago

Great idea, Olga. I just built a function like that and identified three entries with issues: Clark_2002, Lawrence_2015, and Phillips_2013. I see you updated Lawrence_2015, so I'll fix the other ones and then implement the function in QAQC.

coreylawrence commented 5 years ago

So just to be clear. The litter layer notation was intentional in some of these studies. Where did we end up on how we would deal with depth for ISRaD?

The issue is, with the Lawrence template for example, the organic layers were not included in the original study. For that work, depth started at zero corresponding to the organic mineral interface. The organic incubations were added latter and the inverse depth representation was used. By just switching the values in that study, you end up with a case where the organic layers overlap the mineral layers, which is not correct. I also don't think it is correct to "shift" the mineral depths down in the template because then the depth increments are offset from the original study...

A better solution might be to fill new depth range column(s) in ISRaD extra, in order to implement a standard depth notation. Trick is being able to recognize the various depth conventions that are used.

jb388 commented 5 years ago

We have a standard depth convention with the mineral-organic interface = 0, thus there shouldn't be any overlapping depths if the litter layers are represented with negative values. No other depth conventions are allowed EXCEPT if the mineral-organic interface is unknown (e.g. deep peats), and in those cases there is a T/F flag (lyr_depth_all_neg).

olgavinduskova commented 5 years ago

Yes, what I've understood from Alison and what is now also mentioned in the Contribute webpage, we should convert the original conventions to the mineral-organic interface = 0 convention where possible when filling in the template.

greymonroe commented 5 years ago

@jb388 did you add this to QAQC? If so, can we close this issue?

jb388 commented 5 years ago

Yes, I added this two weeks ago.